Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ Recolector de Cienci...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
Recolector de Ciencia Abierta, RECOLECTA
Bachelor thesis . 2023
License: CC BY NC ND
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
UCrea
Bachelor thesis . 2023
License: CC BY NC ND
Data sources: UCrea
versions View all 2 versions
addClaim

Separación gráfica e independencia condicional en redes bayesianas

Graphical separation and conditional independence in bayesian networks
Authors: Merino Iglesias, Daniel;

Separación gráfica e independencia condicional en redes bayesianas

Abstract

Las redes bayesianas y, en general, los modelos gráficos probabilísticos constan de un grafo que recoge relaciones de independencia condicional que se cumplen en la distribución de probabilidad que representan. Esta equivalencia entre el conocimiento que ofrece el grafo y el disponible en la distribución de probabilidad subyacente es lo que abre la puerta a enfoques algorítmicos para el cálculo de consultas probabilísticas con la red bayesiana. En este Trabajo Fin de Grado, se estudiará la relación entre los conceptos de separación gráfica e independencia condicional en redes bayesianas. Para ello, se demostrará la equivalencia entre la factorización de la distribución de probabilidad y el cumplimiento de las propiedades global y local de Márkov por parte de la red bayesiana con respecto a su grafo. La demostración será autocontenida y no asumirá de partida, como suele ser habitual, que la red bayesiana satisface la propiedad local de Márkov. Asimismo, se mostrará la aplicación de los resultados anteriores para la construcción de sistemas inteligentes con un conjunto de datos real. Este conjunto de datos contiene información temporal sobre las interacciones de grupos de estudiantes con una aplicación de escritorio para la consecución colaborativa de un objetivo común, así como las puntuaciones asignadas por su profesor a las soluciones presentadas por los grupos. Se construirá una red bayesiana que permita predecir, en distintos instantes de tiempo, si cada grupo de estudiantes conseguirá puntuaciones altas por su desempeño hasta el momento. La implementación se realizará en R con el paquete bnlearn.

Bayesian networks, and probabilistic graphical models in general, consist of a graph that captures conditional independence relationships present in the probability distribution they represent. This equivalence between the knowledge provided by the graph and the one available in the underlying probability distribution opens the door to algorithmic approaches for computing probabilistic queries with the Bayesian network. In this TFG (acronym of Bachelor’s thesis in Spanish), we will study the relationship between the concepts of graphical separation and conditional independence in Bayesian networks. To do so, we will demonstrate the equivalence between the factorization of the probability distribution and the fulfillment of the global and local Markov properties by the Bayesian network with respect to its graph. The proof will be self-contained and will not assume, as is commonly done, that the Bayesian network satisfies the local Markov property. Furthermore, we will demonstrate the application of the aforementioned results in building intelligent systems using a real dataset. This dataset contains temporal information about the interactions of student groups with a desktop application for the collaborative achievement of a common goal, as well as the scores assigned by their instructor to the solutions presented by the groups. We will construct a Bayesian network that allows us to predict, at different time points, whether each student group will achieve high scores based on their performance up to that moment. The implementation will be carried out in R using the bnlearn package.

Grado en Matemáticas

Country
Spain
Related Organizations
  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
    OpenAIRE UsageCounts
    Usage byUsageCounts
    visibility views 145
    download downloads 47
  • 145
    views
    47
    downloads
    Powered byOpenAIRE UsageCounts
Powered by OpenAIRE graph
Found an issue? Give us feedback
visibility
download
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
views
OpenAIRE UsageCountsViews provided by UsageCounts
downloads
OpenAIRE UsageCountsDownloads provided by UsageCounts
0
Average
Average
Average
145
47
Green
Related to Research communities